CIRF：将思维链分词化为可重用的功能单元，用于大型语言模型的高效潜在推理
CIRF: Tokenizing Chain-of-Thoughts into Reusable Functional Units for Efficient Latent Reasoning in Large Language Models
AI总结：
提出CIRF框架，通过将显式思维链中的语义连贯推理单元映射为离散功能令牌，实现动态序列推理，在数学、符号和常识推理基准上取得优于现有隐式CoT方法的准确率-延迟权衡。
摘要：
隐式思维链通过内化显式理由来降低大型语言模型的推理成本。然而，现有方法通常缺乏与显式理由的对齐以及对示例复杂性的适应性。在这项工作中，我们提出了CIRF（思维链转化为可重用功能单元），一个隐式CoT框架，将推理作为离散功能令牌的动态序列进行。CIRF为显式CoT轨迹中的每个语义连贯推理单元分配一个功能令牌。然后对模型进行微调，以自回归方式生成功能令牌及其可选结果，随后生成最终答案。这种设计将潜在推理与功能单元序列对齐，促进了并行训练、显式理由对齐和自适应推理。在数学、符号和常识推理基准上的大量实验表明，与最先进的隐式CoT方法相比，CIRF提供了有利的准确率-延迟权衡。进一步的分析表明，CIRF构建了独特、可解释的功能令牌，从而带来一致的性能提升。
英文摘要：
Implicit Chain-of-Thought (CoT) reduces the inference cost of large language models by internalizing the explicit rationales. However, existing approaches typically lack alignment with explicit rationales and adaptivity to example complexity. In this work, we propose CIRF (\textit{\underline{C}hain-of-thoughts \underline{I}nto \underline{R}eusable \underline{F}unctional units}), an implicit CoT framework that performs reasoning as a dynamic sequence of discrete functional tokens. CIRF assigns a functional token to each semantically coherent reasoning unit in explicit CoT traces. The model is then fine-tuned to autoregressively generate functional tokens and their optional results, followed by the final answer. This design aligns latent reasoning with a sequence of functional units, facilitating parallel training, explicit rationale alignment, and adaptive reasoning. Extensive experiments on mathematical, symbolic, and commonsense reasoning benchmarks show that CIRF provides a favorable accuracy-latency trade-off compared with state-of-the-art implicit CoT methods. Further analyses demonstrate that CIRF constructs distinct, interpretable functional tokens, leading to consistent performance improvements.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
